Metas sensitive information leaked to employees after engineer seeks AI agent who goes rogue

< />Facebook-parent <a id=” captionrendered=”1″ data-src=”https://etimg.etb2bimg.com/photo/129679365.cms” height=”442″ href=”https://hr.economictimes.indiatimes.com/tag/meta” keywordseo=”Meta” loading=”eager” source=”keywords” src=”https://hr.economictimes.indiatimes.com/images/default.jpg” type=”General” weightage=”20″ width=”590″></img>Meta is reportedly having trouble with rogue <a href=AI agents. According to a report by The Information, a Meta AI agent malfunctioned, exposing sensitive company and user data to employees who did not have permission to access it. Per the report, the employee posted on the company’s internal forum asking for help with a technical question. Another engineer asked the AI agent to help analyze the question. However, the agent posted a response without asking the engineer for permission to share it.

What happened next was concerning. The engineer who asked the question ended up taking actions based on the AI agent’s guidance, which as per the report did not give good advice. The AI agent inadvertently made massive amounts of company and user-related data available to engineers, who were not authorized to access it, for two hours.

Meta confirms its AI agent went rogue

Meta confirmed the incident to The Information, designating the incident a “Sev 1,” which is the second-highest level of severity in the company’s internal system for measuring security issues.

Notably, this is not the first time that rogue AI agents have already posed a problem at Meta. Last month, Summer Yue, a safety and alignment director at Meta Superintelligence, posted how OpenClaw agent deleted her entire inbox, even though she told it to confirm before taking any action.

“Nothing humbles you like telling your OpenClaw “confirm before acting” and watching it speedrun deleting your inbox. I couldn’t stop it from my phone. I had to RUN to my Mac mini like I was defusing a bomb,” Yue’s post then read.

In another news, Meta has announced to shutt down the virtual reality social network for Quest headsets. In a community blog post, Meta noted that the Horizon Worlds app will be removed from the Quest store at the end of March and fully discontinued on VR devices by June 15, 2026. After that date, the platform will continue only as a standalone mobile app, which Meta says will allow each platform to “grow with greater focus.”

Join the community of 2M+ industry professionals.

Subscribe to Newsletter to get latest insights & analysis in your inbox.

All about industry right on your smartphone!

  • Download the App and get the Realtime updates and Save your favourite articles.

Leave a Reply

Your email address will not be published. Required fields are marked *